### File    : corstange-bjmes-codebook.txt
##  Author  : Daniel Corstange (daniel.corstange@gmail.com)
##  Created : Tue 28 Nov 2017 --- 07:53 AM
##  Notes   : codebook for BJMES article

* Notes

  - The notes below summarize the coding for each of the variables
    use in the analysis.
  - These variables have been pre-processed to arrive at these
    encodings.
  - The Arab Barometer (wave 3) variables note the original
    variable number for reference to the original dataset.
  - Datasets exported from SPSS *.sav format, via R, to
    Tab-Separated Files for portability.
  - In the process of exporting, R factors lose their logical
    ordering (sorry!) and come out ordered alphabetically.

* Arab Barometer Questions

  - rid
    + respondent identifier
    + row number in original dataset


  - country
    + respondent's country
    + AB original variable: country
    + values:
      * Algeria
      * Egypt
      * Iraq
      * Jordan
      * Kuwait
      * Lebanon
      * Libya
      * Morocco
      * Palestine
      * Sudan
      * Tunisia
      * Yemen


  - understand
    + understand politics
    + AB original variable: q2185
    + values:
      * 0 Low
      * 1 High


  - interest_4
    + interest in politics
    + AB original variable: q404
    + values:
      * 0 Not interested
      * 1 Slightly interested
      * 2 Interested
      * 3 Very interested


  - interest
    + interest in politics (3-category coding)
    + AB original variable: q404
    + values:
      * 1 Not/Slightly interested
      * 2 Interested
      * 3 Very interested


  - know_fm
    + knows identity of the foreign minister
    + AB original variable: q5221
    + values:
      * 0 Incorrect answer/don't know
      * 1 Correct answer


  - know_governor
    + knows identity of the mayor/governor
    + AB original variable: q5224
    + values:
      * 0 Incorrect answer/don't know
      * 1 Correct answer


  - know
    + knowledge index
    + additive index of know_fm and know_governor
    + values:
      * 0 Low
      * 1 Medium
      * 2 High


  - engaged_index
    + engagement/attentiveness index
    + additive index of understand, interest, and know cut into
      rough thirds, subject to integer constraints
    + values:
      * 1 Low
      * 2 Medium
      * 3 High


  - educ
    + education level
    + AB original variable: q1003, q1003t, q1003yem
    + values:
      * 1 Finished primary school or less
      * 2 Finished intermediate level below secondary school
      * 3 Finished secondary school or more


  - pray_daily
    + performs prayers at least daily
    + AB original variable: q6101
    + values:
      * 0 No
      * 1 Yes


  - quran_weekly
    + reads Quran or Bible at least weekly
    + AB original variable: q6106
    + values:
      * 0 No
      * 1 Yes


  - rel_index
    + religiosity index
    + additive index of pray_daily and quran_weekly
    + values:
      * 0 Low
      * 1 Medium
      * 2 High


* Syrian Questions

  - rid
    + respondent identifier
    + row number in original dataset
      

  - respondent_type
    + was the respondent fully randomized with respect to sex?
    + values:
      * 1 Male; female refused to be interviewed
      * 2 Male; male refused to allow a female to be interviewed
      * 3 Fully randomized


  - district
    + respondent's Lebanese district (caza) of residence
    + values:
      * 1  Akkar
      * 2  Aley
      * 3  Baabda
      * 4  Baalbek
      * 5  Beirut
      * 6  Chouf
      * 7  Metn
      * 8  Minnieh-Dinnieh
      * 9  Nabatieh
      * 10 Saida
      * 11 Tripoli
      * 12 Tyre (Sour)
      * 13 West Bekaa
      * 14 Zahle


  - province
    + respondent's Syrian province (muhafaza) of origin
      * 1  Al-Hasaka
      * 2  Al-Raqqa
      * 3  Al-Suwayda
      * 4  Daraa
      * 5  Deir al-Zour
      * 6  Dimashq (Damascus)
      * 7  Halab (Aleppo)
      * 8  Hama
      * 9  Homs
      * 10 Idlib
      * 11 Latakia
      * 12 Quneitra
      * 13 Rif Dimashq (Rural Damascus)
      * 14 Tartus


  - female
    + respondent's sex
    + values:
      * 0 Male
      * 1 Female


  - age
    + respondent's age
    + values:
      * range between 20 to 65 inclusive


  - age_bins_05
    + respondent's age binned into 5-year increments
    + values:
      * 1 20-24
      * 2 25-29
      * 3 30-34
      * 4 35-39
      * 5 40-44
      * 6 45-49
      * 7 50-54
      * 8 55-59
      * 9 60-65


  - educ_illit
    + is respondent illiterate?
    + values:
      * 0 No
      * 1 Yes


  - educ
    + respondent's education level
    + values:
      * 1 Finished primary school or less
      * 2 Finished intermediate level below secondary school
      * 3 Finished secondary school or more


  - syr_room_density
    + number of family members per (number of bedrooms +1) in Syria
    + values:
      * range between 1/2 and 4 inclusive
        

  - leb_room_density
    + number of family members per (number of bedrooms +1) in Lebanon
    + values:
      * range between 2/3 and 12 inclusive


  - kurdish_speak
    + respondent speaks Kurdish "comfortably" (indicator for Kurds)
    + values:
      * 0 No
      * 1 Yes


  - sect
    + respondent's religious sect
    + values:
      * 1  Alawi
      * 2  Armenian Christian
      * 3  Catholic Christian
      * 4  Druze
      * 5  Ismaili
      * 6  Orthodox Christian
      * 7  Other
      * 8  Other Christian
      * 9  Shia
      * 10 Sunni
      * 11 Syriac Christian


  - sunni
    + is respondent Sunni?
    + values:
      * 0 No
      * 1 Yes


  - minority
    + is respondent a minority (Kurd or non-Sunni)?
    + values:
      * 0 No
      * 1 Yes
        

  - pray_daily
    + performs prayers at least daily
    + values:
      * 0 No
      * 1 Yes


  - quran_weekly
    + reads Quran or Bible at least weekly
    + values:
      * 0 No
      * 1 Yes


  - rel_index
    + religiosity index
    + values:
      * 0 Low
      * 1 Medium
      * 2 High
        

  - relpub_econ
    + importance of religion's role in economic/commercial affairs
    + values:
      * 1 Not important at all
      * 2 Not very important
      * 3 Somewhat important
      * 4 Very important
        

  - relpub_pol
    + importance of religion's role in political/administrative affairs
      * 1 Not important at all
      * 2 Not very important
      * 3 Somewhat important
      * 4 Very important
        

  - relpub_crime
    + importance of religion's role in criminal proceedings
      * 1 Not important at all
      * 2 Not very important
      * 3 Somewhat important
      * 4 Very important
        

  - relpub_family
    + importance of religion's role in marriage and divorce
      * 1 Not important at all
      * 2 Not very important
      * 3 Somewhat important
      * 4 Very important
        

  - relpub_family_bi
    + dichotomized version of relpub_family due to extreme skew
    + values:
      * 0 Somewhat important or less
      * 1 Very important


  - relpub_index
    + additive index of relpub_econ, relpub_pol, relpub_crime
      (each "Somewhat important" or greater)
    + values:
      * 0 Low
      * 1 Medium
      * 3 High

       
  - know_fm
    + knows identity of the foreign minister
    + values:
      * 0 Incorrect answer/don't know
      * 1 Correct answer
        

  - know_const
    + knows constitutional length of Syrian president's term
    + values:
      * 0 Incorrect answer/don't know
      * 1 Correct answer
        

  - know_parl
    + knows how many seats are in Syrian parliament
    + values:
      * 0 Incorrect answer/don't know
      * 1 Correct answer
        

  - know_rev
    + knows where the Arab Spring started
    + values:
      * 0 Incorrect answer/don't know
      * 1 Correct answer
        

  - know
    + additive index of knowledge questions
    + values:
      * 1 Low (2 or fewer correct answers)
      * 2 Medium (3 correct answers)
      * 3 High (4 correct answers)
        

  - interest
    + interest in politics
    + values:
      * 1 Low (Not/little interested)
      * 2 Medium (Somewhat interested)
      * 3 High (Very interested)


  - understand
    + understand politics
    + values:
      * 1 Low
      * 2 High


  - engaged_index
    + additive index of understand, interest, and know cut into
      rough thirds, subject to integer constraints
    + values:
      * 1 Low
      * 2 Medium
      * 3 High
        

  - faction_none
    + respondent does not cite a preference over the factions
    + values:
      * 0 No
      * 1 Yes


  - faction_govt
    + respondent cites government first or second
    + values:
      * 0 No
      * 1 Yes


  - faction_fsa
    + respondent cites the FSA first and does NOT cite Islamists
    + values:
      * 0 No
      * 1 Yes


  - faction_islamist
    + respondent cites any Islamist faction at all
    + values:
      * 0 No
      * 1 Yes


  - faction_islamist_domestic
    + respondent cites dominant Islamists but NOT foreign Islamists
    + values:
      * 0 No
      * 1 Yes
        

  - faction_islamist_foreign
    + respondent cites foreign Islamists at all
    + values:
      * 0 No
      * 1 Yes
        

  - faction_oppn
    + respondent cites either the FSA or the Islamists
    + values:
      * 0 No
      * 1 Yes
        

  - faction_all
    + respondent's preferred faction
    + values:
      * 1 Domestic Islamists
      * 2 Foreign Islamists
      * 3 Government
      * 4 Nationalists (FSA)
      * 5 No Preference (none)
